Variant testing with eBay’s GTC and 30 Day listings

I started a thread in the eBay community about a week ago based on a hunch on GTC(Good Til Cancelled) listings, after having looked at my eBay fees more closely and gritting my teeth at the $0.35 I was being charged every 30 days all listings with this format auto-relisted. While I am aware that GTC listings always auto-relisted at a cost of $0.35 every 30 days, that amount becomes more pronounced as the platform continues to struggle to attract new buyers and one’s sales are trending downwards.

Like many sellers on eBay these days, my sales go through longer periods of stagnation compared to years prior, highly noticeable when I look at my month over month and year over year comparative sales data. Some of these items I sell have no direct competitors on eBay, outside of China. One of the downsides of eBay is one’s listings organically gets buried by newly listed items and it’s much worse for your listing category if sellers of newly listed items have decided to provide buyers with “Free Delivery, “Guaranteed Delivery” and “Free Returns”, eBay’s newly created features to improve the attractiveness of the marketplace for buyers. I have another series of articles coming up shortly that will question the validity of eBay’s “this is what buyers want” hypotheses, but this article is not the place and time for that, so let’s assume for now that eBay knows best.

So, your item gets pushed further down the search results page the longer it remains without sales by newly added items and if it wasn’t a popular item by quantity sold to date, number of watchers, number of views and other criteria Cassini (eBay’s search engine and algorithm) is working with such as “free shipping” and “guaranteed delivery”, that listing may be in trouble of not meeting enough “best match” metrics to maintain high enough visibility for shoppers. Liken it to being a candidate for a staffing agency. If they can’t place you and make money from you after X amount of attempts, they quickly move on to the next candidate that may increase their chances of bringing in some revenue, calling you at a much less frequency than before or worse, not at all. It’s all about the “best match”, like so many areas of our lives.

When this has happened to your listing or one suspect’s something is wrong, can anything be done about it?

On 11/12/2018, I ended a GTC listing that was underperforming and relisted it after 30 minutes as a 30-day listing. Within a combined 8 hour window I had 3 sales. I had also changed the title phrasing and created a duplicative listing about a week before ending and relisting this last GTC item. On the duplicate listing, I lowered the item price by removing free shipping, meaning that the buyer may be attracted by the lower item price but would now have to pay for shipping. The lower item price + shipping did factor in all eBay + PayPal fees, in addition to all associated overhead such as printer ink, paper, tape and mailer costs. That item also had two distinct lengths and colors that the other listing did not.

What I basically did here was create a control “A” and then a variant or “B” to test against, breaking the rules of only making slight changes to a variant and opting for a fuller range of changes that I thought would be attractive to buyers yet make each listing distinct. The variant or “B”, which was the GTC listing I ended and then relisted as a 30-day listing sold 3 items across a single buyer on 11/11/2018 because “combined shipping” was a part of the shipping setup so it was more cost efficient to a buyer than buying two pairs from the control or “A” which offered free shipping but at a cost of $7.99 per item. The variant or “B” sold 3 items across 3 different buyers between 11/11/2018 – 11/12/2018.

Did the buyer see both listings and decide the combined shipping of the variant “B” was their better option? I don’t have analytics or insights into that buyer’s behavior because eBay is not a platform I can have hotjar, mouseflow or other tools installed to record buyer sessions and see their paths and study their shopping behavior. While this would be great insights to have, the purpose of this was more to jump-start the listings by forcing Cassini to see them as newly listed items and have them appear at the top of a search query and get out of the pattern of stagnation the control “A” had seemed to be suffering from for a period of 6 days, the longest window I had recorded with no sales for the particular item in question.

Future considerations

I plan to create a 3rd variant and potentially a 4th and make fewer tweaks than the variant “B” to see if I can more closely find patterns in user preferences between the listings. This is tricky for several reasons. One, we are going into the winter months and the shoes these shoelaces best match with is not the greatest choice for the colder winter months ahead. Secondarily, sales of the models of shoes that these shoelaces in question are truly made for has trended downwards and were already suffering from diminishing popularity throughout most of 2018.

If you’ve done any testing of your own around the topic of GTC listings I’d love to hear from you so be sure to drop a comment below and let me know what you discovered or if you have any direct questions for me about something I may have tried but didn’t note in this article.

Update 11/14/2018 – Another item from variant A has been sold. Rebooting the listings visibility by ending it may have jump-started the process of sales.

Leave a Reply

Your email address will not be published.

This site uses Akismet to reduce spam. Learn how your comment data is processed.